22210
Does Infant Temperament Predict Autistic Traits in Toddlers? Findings from a Prospective Longitudinal Study of Singaporean Toddlers

Saturday, May 14, 2016: 11:30 AM-1:30 PM
Hall A (Baltimore Convention Center)
A. Chew1, C. S. Chee2, D. A. Goh3, D. Gan3, S. B. Lim4, P. Agarwal5, B. Broekman6,7, A. Rifkin-Graboi6, S. M. Saw8, Y. S. Chong9, K. Kwek10, P. D. Gluckman6,11, M. Meaney6,12 and I. Magiati1, (1)Department of Psychology, National University of Singapore, Singapore, Singapore, (2)Child Development Unit, Department of Pediatrics, National University Health System, Singapore, Singapore, Singapore, (3)Department of Psychology, National University of Singapore , now at the Ministry of Social and Family Development, Singapore, Singapore, Singapore, (4)Department of Child Development, KK Women's and Children's Hospital, Singapore, Singapore, (5)Department of Neonatology,, KK Women’s and Children’s Hospital, Singapore, Singapore, (6)Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research (A* STAR), Singapore, Singapore, (7)Department of Psychological Medicine, National University Health System, NATIONAL UNIVERSITY OF SINGAPORE, Singapore, Singapore, (8)Saw Swee Hock School of Public Health, National University of Singapore, National University Health System, Singapore, Singapore, (9)Singapore Institute for Clinical Sciences, Agency for Science, Technology and Research (A* STAR), Singapore; Department of Obstetrics and Gynaecology, Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore, (10)Department of Maternal Fetal Medicine, KK Women’s and Children’s Hospital, Singapore, Singapore, (11)Liggins Institute, University of Auckland, Auckland, New Zealand, (12)Ludmer Centre for Neuroinformatics and Mental Health and Department of Psychiatry, McGill University, Montreal, Canada
Background:  

Limited but emerging evidence suggests that early temperament or temperament trajectories may be different in children who are later diagnosed with ASD. However, most existing research is methodologically limited by cross-sectional studies and small sample sizes and to our knowledge, no study has yet examined temperament in relation to ASD symptoms in infants younger than 6 months. Conceptualizing ASD as the extreme of a normal distribution of autistic traits in the general population allows us to investigate variation in autistic traits and their relationship to infant temperament within larger prospective longitudinal studies.

Objectives:  

The current study examined infant temperament as a predictor of autistic traits in a large community sample of unselected toddlers, using a prospective longitudinal design with multiple time-points. Differences in temperament characteristics were examined both cross-sectionally at each time point and longitudinally in terms of changes in temperament trajectories over time.

Methods:  649 infants were recruited from two major public hospitals in Singapore and followed up at regular intervals from approximately 12 weeks gestation to 3 years of age as part of a large longitudinal study of growth and development (GUSTO; Growing Up in Singapore Towards healthy Outcomes). Caregivers completed the Carey Temperament Scales at 3 (N=645), 9 (N=539), and 18 (N=445) months; and the Quantitative Checklist for Autism in Toddlers (Q-CHAT) at 18 (N=368) and 24 months (N=365). 174 caregivers completed all temperament and autistic trait measures at all timepoints.

Results:  

Correlational analyses showed that rhythmicity, distractibility and adaptability at most earlier time-points were positively associated with later autistic traits, with small to medium effect sizes. In regression analyses, infants’ low distractibility and high threshold were the strongest unique predictors of autistic traits. Low persistence and high threshold were the strongest unique predictors of social autistic traits, as compared to greater persistence which predicted non-social/ behavioral autistic traits. Toddlers with higher autistic traits at 24 months showed a developmental pattern of consistently lower rhythmicity between 3 and 18 months, consistently decreasing persistence and increasing threshold to stimulation from 3 months to 18 months, as compared to their peers with low autistic traits at 24 months who displayed more stable levels of persistence and threshold from 9 to 18 months.

Conclusions:  

This study provides preliminary evidence that at least some temperament differences as early as 3 months are associated with later sub-clinical autistic traits in the general population. Studying infant temperament in the context of autistic traits may provide further insight into how early developmental pathways of those who later go on to have high rates of autistic-like behaviors may deviate from those observed in toddlers with fewer autistic traits. Such knowledge may further and inform our understanding of early development in the extreme end of the continuum of autistic traits.